||P.1301 right column 2nd paragraph: "Second, self-assembly of the flagellar motor appears to constrain the abundance of the protein FliM. In the
motor, FliM forms a ring of ~32 subunits (ref 28) that bind CheY-P to mediate switching of the direction of motor rotation. Studies (refs 21,27,28,31) reveal that only a small
fraction of FliM (<30%) is part of complete motors (see Supporting Material, esp. Table S10)." Supporting information p.17 3rd paragraph: "[Investigators] can also estimate the number of FliM that actually belong to functional flagella. In wild-type E. coli, (primary source 21) reports on average 2.6 flagella per cell, and (ref 57) reports about 3±1.5 flagella per cell. In S. typhimurium, there are about 6-10 flagella per cell (ref 58). Studies of the stoichiometry of the motor report 37±13 FliM copies per motor (primary source 24, ref 59) (in S. typhimurium), and about 32 (primary source 22) (in E. coli).
This would give a number of order 100 FliM that actually belong to functional flagella in E. coli (about 7%), and about 300 in S. typhimurium (about 18%). This is even less than what would be expected from the results above (Table S10), especially for E. coli. This difference may indicate that some complete or almost complete cytoplasmic rings or motors do not belong to flagella. These might be in the last stages of assembly."